In conventional architecture for capturing digital recordings on computers, software waits for notification that audio data is available in a buffer. This results in a particular type of latency in audio recordings that is frequently undesirable. A need exists for a process, preferably a software process, to reduce or eliminate this type of latency in audio recordings.